🔬 An Oscar, Two Asteroids, and the Algorithm in Your sklearn: John Platt on AI for Science
Source
Brandon Anderson
Author
Brandon Anderson
Date
Key takeaways · AI-distilled
ERA (Google's Empirical Research Assistance) treats scientific problems as scoreable tasks: an LLMA large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.Full definition → keeps a tree of past experiment notebooks and picks which to mutate via an Upper Confidence Bound rule, proposing about ten mutations per iteration.
Platt says ERA only started working well after a step change between Gemini 2.0 and 2.5, going from not functioning to functioning great, since this kind of evolutionary search needs the underlying model to actually know where to look.
Using ERA, Platt's team found contrails, the ice-crystal trails from jet exhaust, account for about 1% of human-induced global warming; a single gram of engine exhaust can seed roughly ten kilograms of ice crystals.
The fix for contrails, flying a level or two lower through ice-supersaturated air, was already known, but modeling how much warming it actually prevented stumped Platt's team for two years until ERA found a simpler model accounting for reflected sunlight they had missed.
Platt warns ERA is a power tool that can slice your fingers off: in a Google contrail-detection Kaggle competition, winning entrants exploited a half-pixel labeling error rather than solving the underlying problem, illustrating Goodhart's law in practice.
Terms in this piece · Glossary
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
Why it matters
Google's John Platt describes a repeatable framework, the 'scoreable task,' for turning hard science problems into search targets an AI can maximize, a pattern applied across climate, fusion, and other domains.
